Real-time human pose recognition in parts from single depth images
Top Cited Papers
- 1 January 2013
- journal article
- Published by Association for Computing Machinery (ACM) in Communications of the ACM
- Vol. 56 (1), 116-124
- https://doi.org/10.1145/2398356.2398381
Abstract
We propose a new method to quickly and accurately predict human pose ---the 3D positions of body joints---from a single depth image, without depending on information from preceding frames. Our approach is strongly rooted in current object recognition strategies. By designing an intermediate representation in terms of body parts, the difficult pose estimation problem is transformed into a simpler per-pixel classification problem, for which efficient machine learning techniques exist. By using computer graphics to synthesize a very large dataset of training image pairs, one can train a classifier that estimates body part labels from test images invariant to pose, body shape, clothing, and other irrelevances. Finally, we generate confidence-scored 3D proposals of several body joints by reprojecting the classification result and finding local modes. The system runs in under 5ms on the Xbox 360. Our evaluation shows high accuracy on both synthetic and real test sets, and investigates the effect of several training parameters. We achieve state-of-the-art accuracy in our comparison with related work and demonstrate improved generalization over exact whole-skeleton nearest neighbor matching.This publication has 20 references indexed in Scilit:
- Relevant Feature Selection for Human Pose Estimation and Localization in Cluttered ImagesLecture Notes in Computer Science, 2008
- Implementing Decision Trees and Forests on a GPULecture Notes in Computer Science, 2008
- Vision-based human motion analysis: An overviewComputer Vision and Image Understanding, 2007
- A survey of advances in vision-based human motion capture and analysisComputer Vision and Image Understanding, 2006
- TextonBoost: Joint Appearance, Shape and Context Modeling for Multi-class Object Recognition and SegmentationLecture Notes in Computer Science, 2006
- Tracking loose-limbed peoplePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Object class recognition by unsupervised scale-invariant learningPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Mean shift: a robust approach toward feature space analysisIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- Shape matching and object recognition using shape contextsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- Shape Quantization and Recognition with Randomized TreesNeural Computation, 1997